REGISTERED REPORTS AS A METHOD TO INCREASE CREDIBILITY OF SCIENCE – EXPERIMENTAL STUDY AMONG PSYCHOLOGICAL STUDENTS

Abstract

Aims: The modern scientific publishing system suffers from many problems, amongst which one of the most important is the pressure to publish positive results. A potentially simple way to mitigate publication bias are reviews of manuscript, prior to the data collection and analyses, as well as results presentation and discussion (i.e. Registered Reports). Methods: To test this prediction we conducted a quasi-experiment: two groups of students (n=38), as a part of their academic classes, were asked to design and conduct research projects. They were divided in two groups – Registered Reports- and control group. In both groups students have been encouraged to publish their papers in a local scientific journal. Results: Analyses revealed significant differences in p levels between groups, suggesting more reliable scores for Registered Reports group. Conclusions: Our study partially confirmed the stated hypothesis and suggested, that registered reports might mitigate publication bias. Future recommendations are advised.